Predictive value of a nomogram based on multiparameter quantitative ultrasound for pregnancy outcomes in infertile women attempting natural conception.

Background: Infertility affects many couples worldwide, and the accurate assessment of endometrial receptivity (ER) is critical. However, conventional ultrasound has limited predictive value. Multiparameter quantitative ultrasound is a promising tool for comprehensive evaluation, yet most studies focus on assisted reproduction populations, leaving its combined predictive value for natural conception unclear. This study aimed to explore its value for predicting pregnancy outcomes in infertile women attempting natural conception. Methods: Infertile women who presented to Shenzhen Second People's Hospital from September 2023 to December 2024 were prospectively enrolled. Patients underwent transvaginal ultrasound examination during the mid-luteal phase. Evaluated parameters included endometrial thickness and pattern, direction and frequency of endometrial peristalsis, proportion of endometrial peristalsis duration, uterine artery blood flow characteristics, endometrial and subendometrial blood flow, endometrial volume (EV) and stiffness, and junctional zone (JZ) characteristics (thickness, blood flow, and stiffness). Clinical pregnancy was followed up, dividing patients into pregnancy and non-pregnancy groups. Group comparisons used independent t-test, Mann-Whitney U test, Chi-squared test, or Fisher's exact test as appropriate. Least absolute shrinkage and selection operator (LASSO) regression selected predictors for nomogram construction. Model performance was assessed by receiver operating characteristic (ROC) curve with 500 bootstrap internal validations, calibration curve, and decision curve analysis (DCA). Results: A total of 102 infertile patients were enrolled, of which 22 comprised the pregnancy group and 80 the non-pregnancy group. The results of the study showed a statistically significant difference between both groups in terms of endometrial thickness, direction of endometrial peristalsis, proportion of endometrial peristalsis duration, number of endometrial and subendometrial vessels, endometrial and subendometrial vascular pattern, endometrial and subendometrial blood flow distribution, EV, endometrial vascularization index (VI), endometrial vascularization flow index (VFI), endometrial mean Young's modulus (Emean), JZ-VI, JZ-VFI, EV/JZ volume (EV/JZ-V), and JZ-Emean (P<0.05). LASSO regression selected six important parameters related to pregnancy outcome: proportion of endometrial peristalsis duration, direction of endometrial peristalsis, EV, endometrial Emean, JZ-Emean, and JZ-VFI. The nomogram constructed on the basis of these parameters showed good predictive efficacy, with an area under the curve (AUC) of 0.862 [95% confidence interval (CI): 0.760-0.964], a sensitivity of 0.882, a specificity of 0.758, and an accuracy of 0.783. The calibration curve showed a good agreement (Brier score 0.066, Omnibus P=0.051), and the DCA showed good clinical usefulness. Conclusions: The nomogram, based on the proportion of endometrial peristalsis duration, direction of peristalsis, EV, endometrial Emean, JZ-Emean, and JZ-VFI, demonstrates good predictive performance. By integrating a multidimensional 'structure-function-blood flow' assessment, it provides an objective tool for individualized counseling and optimizing conception timing.